英文.docx

英文.docx

《英文.docx》由会员分享，可在线阅读，更多相关《英文.docx（14页珍藏版）》请在冰豆网上搜索。

英文

DiaphragmSpringClutchDynamicCharacteristicTestBench

M.Velardocchia,G.Ercole,G.MattiazzoandS.Mauro

DepartmentofMechanics,PolitecnicodiTorino

F.Amisano

MagnetiMarelli–EngineControlDivision

Copyright©1999SocietyofAutomotiveEngineers,Inc.

ABSTRACT

Acrucialproblemindesigningaservoactuatedmanualtransmissionisposedbytheclutchtorquevariationswhichdependonoperatingconditionswithequalclutchpedalpositionandclutchslip.Anexperimentalstudyofthephenomenathatdeterminetorquevariationscanbeparticularlyusefulincopingwiththisproblem.Tothisend,theauthorsdevelopedapowertraintestbench.The

benchmakesitpossibletomeasurethemainmagnitudes（disctemperature,clutchslip,clamploadandclutchdiscdisplacement）whichcaninfluencetransmittedtorque.Testbenchlayoutisdescribed.Someexperimentalresultsarepresentedanddiscussed,withparticularattentiontotheproblemsconnectedwithclutchcontrollerdesign.

INTRODUCTION

Diaphragmspringclutchesarecommonlyusedinpassengercarsandlighttruckstocontroldrivelinedynamicsduringmanualsynchromeshgearshifts.Themanualtransmissionclutchsystemdataindicatedbymanufacturerssatisfythemaincustomerneedsregardingtransmissibletorqueandpower,clampload,dimensions,andsoon.Unfortunately,thesedataareincompleteasregardstheextenttowhichclutchsystemperformancedependsoncertainoperatingconditions（e.g.slippingvelocity,shiftingspeed,temperatureinsideclutchhousing,etc.）andcertainclutchparameters（e.g.frictioncoefficientdependenceonslippingandwear）.Thislackofinformationisnotimportantinconventionalsynchromeshtransmissions,wherethedrivercancompensateforchangesintransmittedtorque.Bycontrast,athoroughknowledgeofthesefactorsisneededinordertodesigneffectivecontrollersforservoactuateddrivelineswhichretaintheconventionalpowertrainlayout,butwheretheclutchpedalisreplacedbyaservoactuator.Asthecontrollermustoperateinopenloopconditions,itmustcorrectlyforecasthowclutchtorquevarieswithoperatingconditions（transmissibilityfunction）.Severalstudieshaveshownbysimulationthattheclutchoperatingstrategyusedtoimproveengagementqualitycanchangesignificantly

withclutchtransmissibility[1][2][3].Thisconsiderationhasattractedmanyauthors'attentiontoinvestigatingthetransmissibilityfunction.Thestateoftheartfordiaphragmspringdryclutchsystemsiswelldescribedin[4].Osanai[5],Pacey[6]andWeaver[7]studiedeffectsoftemperatureontransmissibilitycapabilityandheattransferinwetclutches.Zhi-pengHan[8]analyzedthetemperaturedistributiononclutchdiscs.Hiromichi[9]addressedtheproblemofclutchperformanceprediction.

Theaimoftheresearchcarriedoutbytheauthorsistodesignandtestacontrollerforaservoactuatedtransmission.Suchacontrollercanbeconsideredeffectiveifitprovidesperformanceequaltothatachievedbyahighlycompetentdriverwiththesamekindofcarineachpossibleoperatingcondition.Itshouldbeborneinmindthathowoptimalperformanceisdefinedwilldependontherequireddrivingmode,i.e.,sport,comfort,etc.Indesigningacontrollerofthistype,itisnecessarytobeabletocarryoutseveralkindsofexperimentaltests,bothtomeasuretransmittedclutchtorqueandtoassesscontrollerperformance.Thispaperdescribesthetestbenchdevelopedforthispurposeandtheresultsobtainedduringthefirsttests.Thetestbenchmakesitpossibletomeasureshaftspeeds,clampload,pushplatedisplacement,clutchdisctemperature,pedalpositionandtransmittedtorque.Thetransmissionshaftcanbeconnectedbyatorquecontrolledhydraulicmotor（dynamictests）orconstrainedtotheframe（statictests）.

Thepaperpresentstheresultsofaninitialseriesofstatictestscarriedoutwithanearlierversionofthetestbenchwhichwasnotcapableofmeasuringclamploadandactualclutchdiscdisplacement.Uncertaintyininterpretingresultsmadeitnecessarytoimprovethebenchpriortocarryingoutnewtestswhichwillbethefirststepinthefuturedevelopmentoftheresearch.

TESTBENCHDESCRIPTION

Thetestbenchwasdesignedsothatthediaphragmspringclutchcanbetestedwhilesimulatingthemostcommonoperatingconditions,i.e.standingstartsandgearshifting.Thedrivenshaftisthusconnectedtoanhydraulictorquecontrolledmotorusedasabrakeandcontrolledinordertosimulatevehicleinertialload.Thedriveshaftispoweredbya45kWhydraulicmotorwhosespeedcanrange,throughareducer,from30to300rpm.

Amainmotor

（1）isconnectedtoareducer

（2）andfinallytoadriveshaft（3）.Theclutchiscontainedinitshousing（4）,whichalsoprovidesitsdrivingmechanism.Thedrivenshaft（5）islinkedtoabrakingmotor（6）.Thebrakingmotorcanbereplacedwithafixedleverlinkedtotheframeiftestsaretobecarriedoutwiththedrivenshaftrestrained.Theclutchspringisoperatedbyapneumaticcylinderwhichsubstitutesthepedal.

Thebenchisequippedwithsensorstomeasurethemagnitudesthatcandefineandinfluencetheclutchtransmissibilityfunction.Thisfunctionisusuallyconsideredtobetheproductofthefrictioncoefficient,theclamploadandthemeanclutchdiscradius.Thefrictioncoefficient,thoughusuallyconsideredconstant,canactuallydependontemperatureandrelativespeed,whilethemeanradiuscanvarywithdiscwearandtemperature.Sensorswerethusarrangedinordertomeasuretransmittedtorque,relativespeed,actualcontactpressurebetweenclutchdiscandflywheelandtemperatureofthecontactsurfaces.

Thetransmittedtorqueismeasuredbystraingaugesinstalledonthedriveanddrivenshafts,andwhichthusmeasurethetorquetransmittedbyeachshaft.Ifthedrivenshaftisrestrainedbythelever,thetorquecanbemeasuredbyaloadcell.Relativespeediscomputedfromtheshafts’speed,measuredbytwopickupsfacingtheflywheelringgearandatonewheelshrink-fitonthe

transmissionshaftrespectively.

Clamploadcanbemeasuredbystraingaugesinstalledintheshanksofthescrewssecuringtheclutchdisctotheflywheel.Astheseboltsprovidethereactiontodiaphragmspringforce,theirtensionisproportionaltotheforceactingontheclutch.Toensureaccuratemeasurementandreducepossibleerror,threeboltslocated120°apartweresensorized,whilethreeadditionalunloadedsensorizedboltsinstalledinthesamepositionrelativetotheclutchdiscareusedforthermalcompensation.

Contactpressurebetweenthefrictionsurfacesismeasuredbyflatpiezoresistivesensorsinsertedbetweenthetwoclutchdiscs.Theroundsensorsare12mmindiameterand0.1mmthick.Sixsensorsarearrangedonthediscsurfaceatdifferentdistancesfromitscenterinordertoevaluatecontactpressureanddistribution.Unfortunately,thesesensorsshowhighhysteresisandlowaccuracy,andthemeasuredvaluesmustthusbecheckedagainstthetotalforcemeasuredbythesensorizedbolts.However,theycanprovideusefulinformationaboutcontactpressuredistribution.

Flywheelandclutchdisctemperatureismeasuredbythermocouplesplacedinspeciallyproducedseatsdistributedacrossthefrictionsurfaces.Threethermocouplesareinstalledontheclutchdiscatdifferentdistancesfromitscenterinordertoevaluatetemperaturedistributionalongtheradius.Similarly,threethermocouplesarelocatedalongtheradiusoftheflywheel.Afinalthermocoupleisplacedintheclutchhousingtomeasureambienttemperature.

Alaserdisplacementsensormeasuresthepositionofareflectingdiscrigidlyconnectedtothepushplateouterring.Displacementismeasuredrelativetothetransmissioncover.

Duetothelimitednumberofchannelsprovidedbyremotemeasurementhardware,itisnotcurrentlypossibletoacquireallthemeasurablemagnitudesduringeachtest.Accordingly,atestpolicywasselectedwherebytransmittedtorqueisalwaysmeasured,andsignalsregardingclamploadorflywheeltemperatureareacquiredinalternatetests.Similarly,signalsprovidedbycontactpressuresensorsmustbeacquiredinalternationwithsignalsfromclutchdiscthermocouples.Finally,themainmotorsupplypressureandthepositionofthepneumaticcylinderusedtooperatetheclutcharerecorded.

TESTPROCEDURE

Differentproceduresmaybeusedaccordingtotheaimofthetest.Specifically,itispossibletoteststandingstartsorgearshiftoperation.Testscanalsobecarriedoutwiththedrivenshaftconstrained,withthemainadvantageofprovidingawiderrangeofsignalsgiventhatremotemeasurementhardwareisnotneededonthedrivenshaftside.

Testscarriedoutwiththerotatingtransmissionshaftcanbedirectlyrelatedtoactualdrivingconditions,andareparticularlyusefulinstudyingclutchcontrolstrategies.

Bothstandingstartandgearshiftmaneuverscanbesimulated.Inbothcases,thebrakingtorquereferencevalueiscomputedinordertosimulatethevehicleinertialloadandthestaticloadsaccordingtothegearratio.Inparticular,thebrakingmotorcanbeusedasapowermotorfor

simulatingroadslopeduringuphillstandingstarts.

Themainmotortorquecanberegulatedthroughitspressurereducingvalve.

Bycontrast,testscarriedoutwiththeshaftconstrainedareusefulforclutchtransmissibilityfunctionidentification.Unfortunately,testscarriedoutwiththislayoutmaybeaffectedbyerrorsresultingfromthebuild-upofwearmaterialonthefrictionsurface,whichcansignificantlyinfluencethefrictioncoefficient.Whentheclutchdiscrotates,asinactualservice,particlesareejectedbycentrifugalforc